Synthesis of specifically deuterated S-benzylcysteines and of oxytocin

J. Org. Chem. , 1976, 41 (8), pp 1353–1358. DOI: 10.1021/jo00870a014. Publication Date: April 1976. ACS Legacy Archive. Cite this:J. Org. Chem. 41, ...
0 downloads 0 Views 920KB Size
J.Org. Ch,em., Vol. 41, No. 8, 1976

Specifically Deuterated S-Benzylcysteines tetramethylpiperidinyl carbonate, 58229-35-9; l-oxid0-2,2,6,6-tetramethylpiperidinol, 58229-36-0; p-nitrophenyl chloroformate, 7693-46-1.

References and Notes (1) R. B. Merrifield, Adv. Enzymol.. 32, 221 (1969). (2) W. S. Hancock, G. R. Marshall, and P. R. Vagelos, J. Biol. Chem., 248, 2424 (1973). ( 3 ) J. T. Sparrow, A. M. Gotto, Jr., and J. D. Morrisett, Roc. Natl. Acad. Sci. U.S.A., 70, 2124 (1973). (4) H. B. Brewer, Jr., R. Shulman, P. Herbert, R. Ronan, and K. Wehrly, J. Biol. Chem., :!49, 4975 (1974).

1353

(5) E. Bayer, E. Breitmaier, G. Jung, and W. Paar, Hoppe-Seyler’s2. Phys-

io/. Chem., 352, 759 (1971). (6) G. W. Treagear in “Chemistry and Biology of Peptides”, J. Meienhofer, Ed.. Ann Arbor Science Publishers. Ann Arbor. Mich.. 1972, D 175. (7) W.B. Jakoby and M. Wilchek, Methods Enzymol., 34 (1974): ( 8 ) A. M. Felix and M. H. Jimenez, Anal. Biochem., 52, 377 (1973). (9) B. F. Gisin, Helv. Chim. Acta, 58, 1476 (1973). (IO) L. J. Berliner, “Spin Labeling: Theory and Applications”, Academic Press, New York, N.Y., 1975. (1 1) J. M.Stewart and J. D. Young, “Solid Phase Peptide Synthesis”, W. H. Freeman, San Francisco, Calif., 1969, p 27. (12) C. H. W. Hirs, MethodsEnzymol.,11, 361 (1967). (13) L. Chanffe, L. S.Andrews, and R. M. Keefer, J. Org. Chem., 31, 3758 (1966).

Synthesis of Specifically Deuterated S-Benzylcysteines and of Oxytocin and Related Diastereomers Deuterated in the Half-Cystine Positions1g2 Donald A. Upsonl a n d Victor J. Hruby* Department of Chemistry, University of Arizona, Tucson, Arizona 85721 Received August 19,1975 S-Benzylcysteine derivatives specifically deuterated at the a carbon only, the p carbon only, and at both the 01 and p carbons have been.synthesized. These labeled compounds have been enzymatically resolved and the enantiomers and reacemates have been converted to the N-tert- hutyloxycarbonyl derivatives. The deuterium labels were not exchanged under the conditions of the syntheses. Condensation of the sodium salt of difthyl a-acetamidomalonate with benzyl chloromethyl sulfide followed by hydrolysis with DCl afforded S-benzyl-DL-[a-*Hl]cysteine. Acetylation followed by treatment with hog renal acylase separated the stereoisomers. A Mannich reaction with [2H2]methylenediacetate, diethyl a-acetamidomalonate, and dimethylamine followed by quaternization of the amino nitrogen with methyl iodide gave diethyl a-acetamido-a-dimethylamino[2H~]methylmalonate methiodide (15). Treatment of 15 with sodium benzylmercaptide gave diethyl a-acetamid~-a-benzylthio[~H~]methylmalonate, which was hydrolyzed with HCl to yield S-ben~yl-DL-[P,/3-~Hz]cysteine or with DCl to afford S-benzylDL-[a,P,P-2H3]cysteine. These compounds were resolved as before. The preparation of S-benzyl-DL-[a,P,P2H3]cysteine required an efficient source of ethanol-d. This deuterated solvent was prepared in quantitative yield in 2 h from tetraethoxysilane, D20,and a catalytic amount of thionyl chloride. The protected deuterated amino acids were used in the preparation of several oxytocin analogues in which the specific deuteration appears in either the 1-hemicystine or the 6-hemicystine residues. Specific deuterium labels in amino acids, peptides, a n d proteins are very useful for studying the chemical, biological, a n d especially the physical properties of these molecules. If the label is introduced a t nonexchangeable positions, it provides a durable marker which does not significantly alter the properties of the m ~ l e c u l e . ~ - ~ ~ T h e usefulness of such labels has been demonstrated for proteins in studies of structure and folding.14-19 Unambiguous proton8-11220-22and ~ a r b o n - 1 3 ~ nuclear ~ J ~ s ~magnetic ~ resonance assignments for many peptides have been made using specifically deuterated derivatives. T h e development of deuteron magnetic resonance spectroscopy has opened the way for a direct study of the microdynamical behavior of specific segments of the neurohypophyseal peptide hormones a n d their binding t o the neurophysins, their biological carrier protein^.^^,^^ These studies have begun to provide insights into the conformational aspects of these biologically active compounds, giving more direct evidence for structure-activfty relationships t h a n has been available from amino acid substitutions, which are more likely t o perturb the structure of the m o l e ~ u l e . ~ ~ - ~ ~ In order t o perform these experiments it is necessary t o prepare the appropriate specifically deuterium labeled amino acid or amino acid derivative by a synthetic route which maintains the integrity of the deuterium label throughout the synthesis. We report here the total synthesis of S-benzylcysteine deuterated specifically in the a position only, the (3 posi-

tions only, a n d in both the a a n d 0 positions. T h e specifically deuterated cysteine derivatives were also resolved into their enantiomeric pairs without loss of the label, and one was used for the synthesis of the specifically labeled oxytocin [6-hemi[P,P-2H2]cystine]oxytocin(2). T h e race-



H-Cys-Tyr-Ile-Gln-Asn-Cys-Pro-Leu-Gly-NH, 1 2 3 4 5 6 7 8 9 1

mic compounds were used t o incorporate specifically deuterated cysteine into the half-cystine positions of the oxytocin derivatives [6-hemi-DL-[a-2Hl]cystine]oxytocin (3), [1-hemi-DL-[P,(3-2H2]cystine]oxytocin(4), and [ 1-hemiDL-[ol,P,0-2H3]cystine]oxytocin( 5 ) and the diastereomeric pairs were separated by partition chromatography on Sephadex G-25.31,32345 The synthesis of each of the partially deuterated S-benzylcysteine derivatives was accomplished using different approaches. The S-ben~yl-DL-[ol-~Hl]cysteine (6) was made by displacement of chloride from benzyl chloromethyl sulfide by sodium diethyl acetamidomalonate followed by hydrolysis of’the adduct with 11 N DCl in D2O (Scheme

I). T h e S-ben~yl-DL-[@,P-~Hz]cysteine (7) and S-benzyla modification of the methods of Cornelius30 and Atkinson, et al.33 This approach involved making the Mannich adduct of diethyl acetamidomalonate, formaldehyde, or its equivalent, a n d DL- [a,@,/3-2H3]cysteine(8) were made using

1354 J. Org. Chem., Val. 41, No. 8, 1976

Upson and Hruby

Scheme I

+

@CH2-*CH,-Cl

-

PE' I

DMF

Na' -C-NHCOCH, C02Et

11

CO3t

1 -

&H2-s-CHZ+-NHCOCH3

DC1-D 0

I, kO2Et w)

-

D I

Q - C H ~ - - S - - C H , - - C - NI H , +

I

boo 6

Scheme I1 C0,Et

+

(CH3XNH f CDz(0COCH3)2 9

I I C0,Et

H-C-NHCOCHJ

-+

COZEt

(CH3)ZN-CDz-

I --T\"COCH,

(7

CH I 4

C0,Et C02Et

+

@CH:-S-Sa+

(CH&??-CDZ1-

-NHCOCH, kOzEt

f

EtOR R = "or 2H

15

r

C02Et

1

L

k02Et

1 R

boo7,R='H S,R=,H

abstraction from the solvent by the attendant carbanion lead t o the unexpected product. When the displacement reaction was run in ethanol-d (121, however, the same ethanolysis a n d decarboxylation occurred, b u t this time a deuteron was abstracted from the solvent rather t h a n a proton. After DC1 hydrolysis 8 was obtained with >95% deuteration a t both the a and p positions. These experiments required a rapid and inexpensive source of ethanol-d. T h e published34 preparation of ethanol-d from tetraethoxysilane and D20 gave a 90% yield after 24 h reaction and distillation of the product. We modified this approach by the addition of a catalytic amount (ca. 0.3%) of thionyl chloride which produced a slight concentration of D+ in solution.35 We have obtained ethanol-d in 99.9% yield after only 2 h reaction time which was pure as judged by N M R and gas (7) was chromatography. S-Ben~yl-DL-[/3,(3-~Hz]cysteine readily resolved into its enantiomers by N-acetylation followed by reaction with hog renal acylase.36 A similar treatment of 6 and 8 separated the enantiomers without loss of label a t either the a or /3 positions. The resulting S-benzylL-[/3,P-2H2]cysteine (7a) as well as 6, 7, and 8 were readily converted to N-tert-butyloxycarbonyl (N-Boc) derivatives using the general method employing tert- butyl azidoformate.37 No exchange of deuterium was noted in either of the above procedures. The solid-phase synthesis3s of the oxytocin derivatives was carried out on an automated Vega Model 95 synthesizer, a solid-state version of our automated instrument,39 or on a semiautomated apparatus designed and built in our laboratory. T h e methodology used is shown in Table I. Each coupling was >99.4% complete as judged by the ninhydrin test.40 I t was of considerable interest t h a t the active ester couplings of asparagine and glutamine catalyzed by the presence of 1 molar equiv of 1-hydroxybenzotriazole were complete after 3.5 and 5 h, respectively. Without the catalyst, these couplings normally require 8-14 h. Similar catalysis of nitrophenyl ester coupling in solution peptide synthesis has been reported.41 At the conclusion of the synthesis, the N-terminal Boc group was removed and the protected nonapeptide amide was obtained by ammonolysis (in methanol) of the corresponding peptide-resin ester. T h e nonapeptide was deprotected with sodium in liquid ammonia4* and then oxidized under nitrogen43 with 0.01 N K ~ F ~ ( C N )The G . ~isomers ~ were separated from each other and from by-products by partition chromatography on Sephadex G-25 using the solvent system 1-butanol-3.5% At R f 0.32, the HOAc in 1.5% aqueous pyridine (l:l).45 peak for the hemi-D-cysteine derivatives of oxytocin appeared, and a t Rf 0.24, the peak for the oxytocin derivatives appeared, both R f values in close agreement with our previous experience with diastereomeric mixtures.45 T h e diastereomers were isolated and further purified by gel filtration on Sephadex G-25 using 0.2 N HOAC as eluting solvent. T h e purity of the products was checked by thin layer chromatography in three solvent systems, quantitative amino acid analysis, and optical rotation determination. T h e milk ejecting activitie@ of the partially deuterated oxytocin derivatives were indistinguishable within experimental error from those of the native hormone. T h e partially deuterated hemi-D-cystine derivatives all had milk ejecting activities greatly reduced compared t o those of oxytocin. These derivatives have been used for the assignment of a,P and carbonyl carbons in 13C N M R s p e ~ t r a , ~ ~ , ~ ~ and except for these carbon atoms t h e 13C N M R spectra, were indistinguishable from those of authentic oxytocin.

dimethylamine. T h e adduct was quaternized with methyl iodide, and the product was treated with sodium benzylmercaptide. Upon hydrolysis, the resulting product gave S-benzyl-DL-cysteine (Scheme 11). In place of formaldehyde we used [~y,a-~H2]methylene diacetate (9). Scrupulously anhydrous conditions were required before 9 could be isolated in good yield. Since t h e Mannich adduct is deuterated, one expects t o obtain 7 when HC1 is used in the hydrolysis step, and 8 when DC1 is used. This approach gave 7 as expected with >95% deuteration a t the /3 positions. However, when the synthesis of 8 was attempted in this manner, we obtained the S-benzylDL-cysteine derivative which was >90% 7 (virtually undeuterated in the a position). The reaction of sodium benzylmercaptide with the quaternary salt was done for 6 days in refluxing ethanol. Isolation and characterization (NMR) of the reaction product revealed t h a t it was not the expected Experimental Section adduct b u t was ethyl N-a~etyl-S-benzyl-DL-[/3,/3-~H~]cysteinate (>go%). Apparently, during the course of the reacThin layer chromatography (TLC) was done on silica gel G plates using the following solvent systems: (A) 1-butanol-acetic tion, ethanolysis, decarboxylation, and subsequent proton

J . Org. Chem., Vol. 41, No. 8, 1976 1355

Specifically Deu.terated S-Benzylcysteines

Table I. Solid-Phase Peptide Syntheeis Methodology Nitrophenyl ester coupling (for Asn and Gln only)

Normal DCC Coupling Time, min

No. of times

Step

Solvent or reagent

1

1

2

4 1 1

2 3

20

1

4

1

3 2 4 1

5 6 7 8

CHzClz Ninhydrin test TFA-CHzClpanisole (25:73:2) TFA-CHzClz-anisole (25:73:2) CHzClz DIEA-CHzClz (10:90) CHzClz DMF

20

1

9

1 CHZCl2 1 100%EtOH 1 12 CHZC12 Amino acid (1.5 equiv)13 CHZClZ 20 14 DCC (1.2 equiv)CHzClz 1 15 CH2Clz 1 16 100%EtOH a All amino acids are Ne-Boc protected.

2 2 3

10

1

12

Step 1

2 3 4 5 6 7 8

9 10 11

Solvent or reagent CHzClz Ninhydrin test TFA-CHzClz-anisole (26:73:2) TFA-CHzClz-anisole (25:73:2) CHzClz DIEA-CH2C12 (1090) CHZC12 Amino acid" (1.5 equiv)-CHzClz DCCl (1.2 equiv)CHzClz

2 1

11

Amino acida (4 equiv)DMF, 1-Hydroxybenzotriazole ,(4equiv) DMF CHzClz 100%EtOH

Time, min

No. of times

1

4

2

1 1

20

1

1

1 1

3 2 4 5

210-300

1

1 1 1

3 2 2

2

1

2 2

acid-water (4:1:5, upper phase only); (B) 1-butanol-acetic acidpyridine-water (15:3:1012); (C) 1-pentanol-pyridine-water (35: 3530). Capillary melting points were determined on a ThomasHoover melting point apparatus and are uncorrected. Nuclear magnetic resonance (NMR) spectra were obtained using a Varian T-60 NMR spectrometer or a Bruker WH-90 NMR spectrometer. Amino acid analyses were obtained by the method of Spackman, Stein, and Moore48 on a Beckman 120 C amino acid analyzer after hydrolysis in 6 N HC1 for 22 h. Optical rotation values were measured a t the mercury green line (547 nm) using a Zeiss Old 4 polarimeter, Elemental analyses were performed by Spang Microanalytical Laboratory or Heterocyclic Chemical Corp., and deuterium analyses were performed by Joseph Nemeth, Urbana, Ill. Benzyl Chloromethyl Sulfide (11). The title compound was made in 59% yield following the reported p r 0 c e d ~ i - ebp : ~ ~102 "C (2 mm) [lit. 102 "C (2 mm)]; NMR (neat) 6 3.55 (s, 2 H), 4.10 (s, 2 H), 7.00 (s, 5 H). Diethyl a-Acetamido-a-benzylthiomethylmalonate(10). A 1.73-g portion of sodium hydride (50% in oil, 0.036 mol) was placed in a dry, nitrogen.filled reaction flask, covered with dry dimethylformamide (DMF), and cooled to 0 "C. A solution of 7.83 g (0.036 mol) of diethyl a-acetamidomalonate in 40 ml of DMF was added to the cold hydride mixture over 20 min. Stirring was continued until hydrogen evolution ceased (ca. 1h). Benzyl chloromethyl sulfide (6.15 g, 0.036 mol) was added, and the flask was placed in a 90-100 "C bath for 66 h. The mixture was cooled, filtered, and concentrated in vacuo to a brown oil. Trituration with 20 ml of water was followed by three 5-ml chloroform extractions. The combined organic layers were dried (KzC03) and filtered, and the filtrate concentrated in vacuo to a brown solid which was treated eight times with 20-ml portions of boiling hexane. Each time the clear hexane layer was decanted from an oily lower layer. As the hexane cooled, white needles were deposited yield 6.68 g (53% of theory); mp 87.3-88 "C; NMR (CDCl3) 6 1.25 (t, 6 H), 2.05 (s, 3 H), 3.55 (s, 2 H),3.75 (s, 2 H), 4.25 (q, 4 H), 6.95 (s, broad, 1 H), 7.30 (6, 5 H). Anal. Calcd for C17Hz3NS: C, 57.79; H, 6.53; N, 3.98. Found: C, 57.35; H, 6.54; N, 4.13. S-Benzyl-DL-[r~-~H~]cysteine (6). A 6.0-g portion of 10 was treated with ca. 11 M DC1 in DzO (made by adding 13 ml of SOClz dropwise to 31.3 in1 of cold DzO) a t reflux for 6 h. After cooling slightly, Norit was added and the mixture was briefly brought t o boiling. The mixture was rapidly filtered and the filtrate was evaporated to dryness in vacuo. The residue was taken up in about 25 ml of water, and concentrated NH40H was added to pH 5.5. After filtration of the resulting crystals and drying over NaOH in vacuo, there was obtained 2.33 g (65% of theory) of 6: mp 215.5-217 "C

(lit. for protio c0mpound,4~215-216 "C); NMR (CF3COOH) 6 2.75 (s, 2 H), 3.40 (s,2 H), 3.60-3.80 (a-CH, undetectable), 6.90 (s, 5 H). N-B~c-S-benzyl-DL-[a-~H~]cysteine (13). The title compound was made in 70% yield following the reported procedure,37 mp 110.5-111.0 "C (lit.45 111-111.3 "c, lit.37 63-65 "C). Single, uniform spots were obtained on TLC using systems A, B, and C with Rf values identical with those of the protio analog: NMR (CDC13) 6 1.45 (s, 9 H), 2.90 ( ~ , H), 2 3.75 (s, 2 H), 4.40-4.60 (a-CH, undetectable), 5.35 (broad, 1 H), 7.30 (s, 5 H). Quantitative deuterium analysis indicated 96% deuteration (calcd 4.75 atom %, found 4.55 atom %). S-Benzyl-~-[a-~H~]cysteine (sa).The resolution of 6 was done using the reported procedure for the protio compound36 with minor modifications. The starting material, N-acetyl-S-benzylDL-[~x-~H,]cysteine, was obtained by adding 1.25 g of 6 to 3 ml of HzO and treating the resulting slur.ry with 0.9 g of N a ~ C 0 3and 1.15 ml of acetic anhydride. After 20 min, the homogenous mixture was acidified with concentrated.HC1 to pH 1. There was obtained 1.42 g (95%), mp 154.5-155.2 "C (lit.36for protio analog, mp 157 "C). The acetylated amino acid was slurried in 90 ml of water and the pH was adjusted to 7.5 with NH40H. The total volume was brought to 110 ml and acylase I (hog kidney, Calbiochem, 0.23 g) was added and the mixture was stirred at 38 "C for 2 days maintaining the pH. An additional 57 mg of enzyme was added and the mixture stirred 2 days more. The pH was lowered to 5 with acetic acid, a few milliliters of butanol were added and the mixture was concentrated to about 20 ml. The mixture was filtered and the filtrate was acidified with concentrated HC1 to give 640 mg of N-acetyl-S-ben~yl-D-[a-~H~]cysteine (90.5%), mp 140-141 "C. The residue on the filter was transferred to a flask where it was boiled with 1 N HC1 and filtered. This process was repeated and the combined filtrates were neutralized with NH40H, giving 390 mg of S-benzylL-[a-2Hl]cysteine (64%), mp 215-218 "C, [a]& -17.4" (c 1, 5 N HC1) [lit.36 [aIz5D -19.5" (c 1, 5 N HCl), commercial amino acid (Fox 4978) [alii7 -17.1'1. A quantitative deuterium analysis showed the compound to be 97% deuterated (calcd 7.69 atom %, found 7.52 atom %). D i b r o m ~ [ ~ H ~ ] m e t h a(14).2s ne Dibromomethane (55 g) was successively exchanged with protions of 10% NaOD in DzO by rapidly stirring at reflux for 24-h periods. After five exchanges using 22-, 13-, 12-, 11-, and 7-ml portions of base, the exchange was judged to be about 80% complete by NMR. Three additional exchanges employing 12 ml each of 10% NaOD in DzO gave 20.2 g of 14 which was >95% deuterated. [a,cu-2H~]MethyleneDiacetate (9).33A 20-g portion of 14 was mixed with 66 ml of anhydrous acetic acid (distilled from triacetyl-

1356 J. Org. Chem., Val. 41, No. 8,1976

Upson and Hruby

borate), 6.7 ml of acetic anhydride (freshly distilled after standing ethyl N-a~etyl-S-benzyl-DL-[a,&@-~H3]cysteinate, which was then over P205), and 32.2 g of anhydrous potassium acetate under dry treated with 11 N DCl in D2O to give 5.55 g of 8 (81%from 15). nitrogen in a flask equipped with a mechanical stirrer. After all the The white crystals melted a t 211-212 O C ; NMR (CF3COOH) 6 2.75 reagents were added, the nitrogen inlet was replaced with a con(/3,/3-CH2,undetectable), 3.40 (s, 2 H), 3.60-3.80 (a-CH, undetectdenser, and the mixture was refluxed for 24 h. Stirring was continable), 6.90 (s, 5 H). ued while the mixture cooled to room temperature. Ether (200 ml) N-Bo~-S-benzyl-DL-[a,~,~-~H~]cysteine (18). A 2.50-g porwas added and the mixture was filtered. The filter cake was tion (0.0117 mol) of 8 was converted to 3.45 g of 18 (94%of theory) washed with four 50-ml portions of ether. The ether and acetic using the same method as used to convert 6 to 13. The white crysacid were distilled from the mixture at atmospheric pressure. At tals melted at 111.5-111.8 "C; NMR (CDC13) 6 1.50 (s, 9 H), 2.90 54-55 "C (10 mm) [lit.33 61-63 "C (12 mm)] there was obtained (P,P-CHZ,undetectable), 3.75 (s, 2 H), 4.40-4.60 (a-CH, undetect7.63 g (49%of theory) of 9 which was >95% deuterated. able), 5.35 (broad, 1HI, 7.30 (s,5 H). A single spot was obtained on Diethyl a-Acetamido-o-dimethylamino[2H~]methylmalon- TLC using systems A, B, and C, with Rf values identical with those ate Methiodide (15). An 11.2-g portion of 9 (0.083 mol) was cooled of the protio analogue. Quantitative deuterium analysis indicated to -10 "C and 27.8 ml of a 40% aqueous solution of dimethylamine 95% deuteration (calcd 14.28 atom %, found 13.50 atom %). was added dropwise such that the temperature did not exceed 0 Boc-Leucylglycinate-Resin (19). A 5.00-g portion of Boc-gly"C. Then 14.9 g (0.083 mol) of diethyl a-acetamidomalonate was cinate-resin (polystyrene crosslinked with 1%divinylbenzene, LS added in one portion. The flask was sealed and the mixture was 601 Merrifield Resin, Lab Systems, Inc., San Mateo, Calif.) substistirred a t 40 "C for 1 h. The flask was cooled to -10 "C and 20% tuted with Boc-glycine a t the level of 0.64 mmol/g by the method aqueous NaOH was added to pH 11.The cold mixture was extractof Gisensl was deprotected and neutralized as described in steps ed with two 50-ml portions of ether. The combined organic ex1-7 of Table I. Limited coupling was done with 0.585 g (2.35 mmol) tracts were dried over Na2S04, filtered, and freed of solvent giving of Boc-leucine-H2O and an equivalent molar quantity of dicycloa semisolid residue. The residue was taken up in 150 ml of dry hexylcarbodiimide (DCC). The reaction was allowed to proceed for ether and 25 ml of methyl iodide was added. The mixture was 30 min. After the appropriate washes (see Table I, steps 10-12), stirred at 40 "C for 24 h, and then allowed to stand at room temthe unreacted amino groups were treated with 0.35 g (3.2 mmol) of perature for 24 h. The white product was filtered and dried in N-acetylimidazole in CHzClz for 3 h. At the conclusion, a ninhyvacuo to give 24.24 g (70% of theory) of 1 5 mp 174-175 OC (lit.30 drin test was negative. The average of four modified52 aldimine53 determinations gave a leucyl substitution level of 0.495 mmol/g. 171-173 "C); NMR (DzO) 6 1.20 (t, 6 H), 2.10 ( 8 , 3 H), 3.15 ( ~ , H), 9 4.35 (q,4 H), 4.35 (s, absent). Solid-Phase Synthesis of S-3,4-Dimethylbenzylcysteinyl0-benzyltyrosylisoleucylglutaminylasparaginyl-SbenzylS-Ben~yl-DL-[B,B-~Hz]cysteine (7). Under dry nitrogen, 0.32 g DL-[a-2H~]~ysteinylpr~lylleucylglycinate-Resin (20). The (0.014 mol) of sodium was added to 30 ml of absolute ethanol. solid-phase synthesis of 20 was performed as a Vega Model 95 synWhen all the sodium had reacted, 1.74 g (0.014 mol) of benzyl merthesizer, an automated machine similar to that described by captan was added followed by 5.84 g (0.014 mol) of 15. The mixHruby et al?9 The preparation was done on a 1.5-mmol scale, ture was refluxed for 6 days, cooled, and freed of solvent. The resiwhich required 3.19 g of the resin 19. The synthetic cycles are outdue was disolved in 20 ml of chloroform and 10 ml of cold water lined in Table I. All reactions and washes were carried out with was added. The mixture was shaken and separated, and the organ30-ml portions. After the last coupling, the terminal Boc group was ic layer was dried over anhydrous KzC03. The drying agent was filremoved (Table I, steps 1-6). The peptide-resin was filtered and tered off and the solvent removed leaving a residue which solididried in vacuo and found to have increased in weight by 1.42 g fied on standing. Without further purification, the residue was (85% of theory). The peptide-resin 20 was ammonolyzed and the treated with concentrated HC1 at reflux for 5.5 h. The mixture was peptide was extracted into DMF and precipitated with water as cooled, treated with Norit, and again boiled. The mixture was fildescribed el~ewhere.4~ There was obtained 1.44 g (74% of theory) tered hot and the filtrate was evaporated to dryness in vacuo. The of the protected nonapeptide amide H-Cys(DMB)-Tyr(Bz1)-Ileresidue was dissolved in about 20 ml of HzO and the pH was adGln-Asn-~~-[a-~H1]Cys(Bzl)-Pro-Leu-Gly-NHz (21), mp 210-214 justed to 5.5 with NH40H. After filtering and drying, there was OC. Anal. Calcd for C,j,jHsoN12012Sy4H20: c , 57.47; H, 6.82. obtained 2.14 g (72% from 15) of white crystals melting at 209-211 Found: C, 57.30; H, 6.95. O C : NMR (CF3COOH) 6 2.75 (P,P-CHz, undetectable), 3.20 (s, 2 [6-Hemi-DL-[a-2H~]cystine]oxytocin (3). The protecting H), 3.40-3.60 (1 H), 6.80 (s, 5 H). Quantitative deuterium analysis groups were removed by treatment of 325 mg (0.25 mmol) of 21 indicated 97% deuteration (calcd 15.40 atom %, found 14.95 atom with sodium in liquid ammonia (freshly distilled from sodium) and %). the sulfhydryl groups were oxidized under nitrogen with 50 ml of N-Boc-S-ben~yl-DL-[~,~-~Hz]cysteine (16). Employing the 0.01 N K3Fe(CN)6 solution.44The product [6-hemi-~L-[a-~H1]cyssame method used to convert 6 to 13, 3.00 g (0.014 mol) of 7 gave tine]oxytocin (3) was purified by partition chromatography using 3.80 g of 16 (87%). The white crystals melted a t 110.5-111.0 OC; the solvent system 1-butanol-3.5% aqueous acetic acid in 1.5%pyrNMR (CDC13)6 1.45 (s, 9 H), 2.90 (P,P-CH2,undetectable), 3.75 ( 8 , idine (1:1).45 Analysis by the Folin-Lowry method54 showed a 2 H), 4.40-4.60 (1H), 5.35 (broad, 1H), 7.25 (s, 5 H). A single spot small by-product peak at Rj 0.61, and a large, poorly resolved peak was obtained in TLC using systems A, B, and C with Rf values at Rj 0.38-0.20 of the incompletely separated diastereomeric oxyidentical with those of the protio analogue. tocin derivatives. The fractions corresponding to the diastereomerS-Benzyl-L-[P,B-2Hz]cysteine(7a). A sample of 7 was acetyic mixture were pooled and the peptide mixture obtained after isolated and resolved with hog renal acylase as described above for lation was subjected to the same solvent system as above. As found compound 6. The resolved L isomer (7a) had mp 215-218 "C [a]& previ0usly4~the diastereomers were nicely separated with peaks a t -20.7' ( C 1 , 5 N HCl). Rj 0.32 and 0.24 corresponding to [6-hemi-D-[a-2H1]cystine]oxytoN-Boc-S-benzyl-~-[fi,~-~H2]cysteine (17). Employing the cin (3b) and [6-hemi[a-2H1]cystine]oxytocin(3a), respectively. same method used to convert 6 to 13,0.75 g (0.0025 mol) of 7 was The oxytocin peak (Rf 0.24) was isolated to give 70 mg of unconconverted to 1.01 g of 17 (92%) which remained an oil: NMR taminated product. The 6-hemi-~oxytocin derivative as isolated (CDC13) 6 1.50 (s, 9 H), 2.90 (/3,P-CH2, undetectable), 3.75 (s, 2 H), contained small amounts of oxytocin. Repurification by partition 4.40-4.60 (1 H), 5.40 (broad, 1 H), 7.30 (s, 5 H), 11.60 (s, 1 H); CY]^?! -48.8O (c 1.0, CH&OOH), for all-protio analogue (Biosyn- chromatography gave 51.4 mg of the pure 6-hemi-D diastereomer 3b. Both 3a and 3b were separately purified by gel filtration chrothetica 6145) [a]f:f -52.2O (c 1.0, CH3COOH). A single spot was matography on Sephadex G-25 (200-270 mesh) using 0.2 N acetic obtained on TLC using systems A, B, and C, with Rj values identiacid as eluent solvent. Final purified yields were 64 mg of 3a and cal with those of the protio analogue. 50 mg of 3b. Each of the isomers gave single spots on TLC in solEthanol-d (12). Under a nitrogen atmosphere, 104.2 g (0.50 vent systems A, B, and C, identical with those of authentic protio mol) of tetraethoxysilane was mixed with 40 g (2.0 mol) of DzO and analogues. Diastereomer 3a exhibited a carbon-13 spectrum idena catalytic amount of thionyl chloride (ca. 0.5 ml). After stirring a t tical with that of authentic oxytocin except for the absence of the room temperature for 2 h, the mixture was distilled at 2 mm. The peak corresponding to the 6-hemicystine a carbon. It had [a];& last traces of product were obtained by increasing the vacuum and -22.6O (c 0.5, 1 N HOAc). A sample of 3a was hydrolyzed in 6 N warming the distillation flask to about 30 "C. There was obtained HC1 for 24 h at 110 "C. Amino acid analysis gave the following 93.9 g (99.9%of theory) of 12. No starting materials contaminated molar ratios: aspartic acid, 1.0; glutamic acid, 1.1;proline, 0.9; glythe product as ascertained by gas chromatography and NMR. The cine, 1.0; half-cystine, 1.9; isoleucine, 1.0; leucine, 1.1;tyrosine, 0.9. infrared spectrum was consistent with spectra of other deuterated Milk-ejecting activity determined using mouse-mammary tissue in alcohols.50 vitro46was identical with that of authentic oxytocin within experiS-Benzyl-~~-[a,fl,j3-~H&ysteine (8). The reaction of 13.30 g mental error. (0.032 mol) of 15 with sodium benzylmercaptide in ethanol-d gave

Specifically Deuterated S-Benzylcysteines

J.Org. Chem., Vol. 41, No. 8, 1976

1357

single, uniform spots identical with those of the protio analogue. Diastereomer Bb had [a]%9,-108,l' (c 0.5, 1 N HOAc) [lit? Milk-ejecting activity was identical with that of authentic oxytocin -81' (c oh, 1N HOAc)], Amino acid analysis gave the fo1within experimental error. lowing molar ratios: aspartic acid, 1.0; glutamic acid, 1.0; proline, A sample of [l-hemi-D-[/3,~-2H~]cystine]oxytocin had [a]% 1.0; glycine, 1.0; half-cystine, 2.0; isoleucine, 0.9; leucine, 1.0; tyro-68.9O (c 0.5, 1 N HOAc) [lit?2 [CX]~OD -56' (c 0.5, 1 N HOAc)]. sine, 0.9. Milk-ejecting activity was ca. 34.7 units/mg. Anal. Calcd Amino acid analysis following hydrolysis for 24 h at 110 "C in 6 N for C ~ ~ H ~ ~ D N ~ ~ O ~ Z S ~ . C, C H50.75; ~ C O ZH,H 6.39; : N, 15.79. HCl gave the following molar ratios: aspartic acid, 1.0; glutamic Found: C, 50.78; II,6.37; N, 16.19. Solid-Phase Synthesis of [6-Hemi[~,~-2H~]cysthe]ox~tocinacid, 1.0; proline, 1.0; glycine, 1.0; half-cystine, 2.0; isoleucine, 1.0; leucine, 1.0; tyrosine, 0.9. Development on TLC in systems A, B, (2). Boc-leucylglycinate-resin with a substitution level of 0.27 and C gave single, uniform spots identical with those of the protio mmol/g was prepared in an analogous manner to that described for analogue. Milk-ejecting activity was ca. 38.6 units/mg. the preparation of 19. A 3.80-g portion of this resin was used to [ l-Hemi-DL-[a,~,~-2H~]cystine]oxytocin (5) and Separation synthesize the title compound on a 1-mmol scale. A semiautomatof the Diastereomers. The same process described for the prepaed apparatus designed and built in our laboratory was used which ration of 4 was repeated, substituting 18 for 16. There was oballowed the reaction vessel to be filled and emptied by application tained 4.87 g of nonapeptide-resin which after ammonolysis, exof vacuum a t the top and bottom, respectively, of the vessel, and to traction, and precipitation g&e 1.24 g of the protected nonapepbe shaken mechanically. For each step, 35 ml of solvent was used tide amide, mp 227-228 "c. Anal. Calcd for C S ~ H ~ ~ D ~ N I ~ O I ~ S Z * except for the DIEA neutralization step in which case 40 ml was 2Hz0: C, 58.86; H, 7.26; N, 12.48. Found: C, 58.82; H, 6.86; N, used. For this synthesis 0.8 equiv of DCC was used for each equiv12.49. A 325-mg sample was deprotected and oxidized as described alent of Boc-amino acid.j6 The N-terminal Boc group was refor compound 21. After the usual chromatographic purifications, moved. The resin was dried and found to have increased in weight there was obtained 77.5 mg of [l-hemi-~-[a,/3,/3-~H~]cystine]oxytoby 1.08 g (105%).The protected nonapeptide was cleaved from the cin (5a) and 72.2 mg of [l-hemi-D-[~~,~,/3-~H~]cystine]oxytocin (5b). resin by ammonolysis. After extraction into DMF, precipitation, Compound 5a had [a]:& -24.5" ( c 0.5, 1 N HOAc). A sample of 5a filtering, and drying, there was obtained 725 mg of nonapeptide, was hydrolyzed in 6 N HC1 for 24 h at 110 'C. Amino acid analysis mp 215-217 "C, and a second crop weighing 390 mg, mp 173-178 gave the following molar ratios: aspartic acid, 1.0; glutamic acid, "C. A portion of the first crop (325 mg, 0.25 mmol) was deprotect1.1; proline, 1.0; glycine, 1.1; half-cystine, 2.0; isoleucine, 1.0; leued with sodium in liquid ammonia and oxidized with 0.01 N aquecine, 1.0; tyrosine, 0.88. On TLC, 5a gave single, uniform spots in ous KsFe(CN)S. Purification by partition chromatography using systems A, B, and C. Milk-ejection activity was identical with that the solvent system 1-butanol-3.5% aqueous acetic acid in 1.5%pyrof authentic oxytocin within experimental error. idine (1:l) gave a peak centered at Rf 0.23 corresponding to the Compound 5b had [ C X ] ~ & -68.4' (c 0.5, 1 N HOAc). Amino acid title compound 2. The fractions corresponding to the peak were analysis gave the following molar ratios: aspartic acid, 1.0; glutampooled and lyophilized. Final purification by gel filtration chromaic acid, 1.0; proline, 1.1; glycine, 1.0; half-cystine, 1.9; isoleucine, tograph afforde'd 63 mg of [6-hemi[/3,/3-2H~]cystine]oxytocin. It 0.9; leucine, 1.0; tyrosine, 1.0. Milk-ejecting activity was ca. 32.6 had [a]& -23.7* (c 0.5, 1 N HOAc). Analysis on TLC in solvent units/mg. systems A, B, and C gave single spots identical with those of an authentic protio analogue. A sample was hydrolyzed in 6 N HCl for 24 h at 110 OC. Amino acid analysis gave the following molar raAcknowledgments. W e t h a n k Dr. M a c E. Hadley, Detios: aspartic acid, 1.0; glutamic acid, 1.0; proline, 1.1; glycine, 1.0; partment of Cell and Developmental Biology, University of half-cystine, 2.0; isoleucine, 0.9; leucine, 1.0; tyrosine, 0.9. MilkArizona, for performing t h e milk-ejecting assays, a n d Mr. ejecting activity was identical with that of authentic oxytocin withDavid E. Wright for performing amino acid analyses. in experimental error. Solitl-Phase Synthesis of Boc- O-benzyltyrosylisoleucylglutaminylasparaginyl-S-3,4-dimethylbenzylcysteinylprolylleuRegistry No.-2, 57866-58-7; 3,57866-59-8; 3a, 57866-60-1; 3b, cylglycinate-Resin (22). Boc-leucylglycinate-resin with a substi57866-61-2; 4, 57866-62-3; 4a, 57866-63-4; 4b, 57866-64-5; 5, tution level of 0.36 mmollg was prepared in an analogous manner 51548-05-1; 5a, 51493-96-0; 5b, 51548-06-2; 6, 57866-70-3; 6a, to that described for the preparation of 19. A 7.50-g portion of this 57866-71-4; 7, 57866-72-5; 7a, 57866-73-6; 8, 51494-04-3; 9, 22117resin was used to synthesize the title compound on a 2.7-mmol 87-9; 10, 57866-74-7; 11, 3970-13-6; 13, 57866-75-8; 14, 22117-86-8; scale using the somiautomated apparatus described above in the 15, 57866-76-9; 16, 57866-77-0; 17, 57866-78-1; 18, 51493-98-2; 19 preparation of 2. In each step 50 ml of solvent was used except for resin free, 32991-17-6; 21,57866-68-9; 22 resin free, 57866-69-0; dithe DIEA neutralization step in which case 60 ml was used. Again, ethyl a-acetamidomalonate, 1068-90-2; N-acetyl-S-benzyl-DL-[cu0.8 equiv of DCC was used for each equivalent of Boc-amino 2H~]cysteine, 57866-79-2; N-a~etyl-S-benzyl-D-[cu-~H~]cysteine, acid.56 After each coupling series, a ninhydrin test indicated 57866-80-5. >99.4% coupling for all but Boc-0-benzyltyrosine; in this case a third coupling using 0.5 equiv was performed, but the reaction was References and Notes only about 94% complete as judged by the ninhydrin test. The unreacted amino groups were acetyl terminated with N-acetylimidaz(1) Financial support from the National Sclence Foundation (GB 40106) and the U.S. Public Health Service (AM 17420) is gratefully acknowledged. ole in CHzClz. The peptide-resin was filtered and dried in vacuo D.A.U. is the recipient of a Lubrizol Foundation Scholarship. and found to have increased in weight by 2.89 g. (2) All amlno acids except glycine are of the L configuration, unless other[ l-Hemi-DL-[~,@-2H~]cystine]oxytocin (4) and Separation wise noted. Standard abbreviations for amino acids, protecting groups, of the Diastereomers. A 4.75-g portion of the Boc-octapeptide and peptides as recommended by the IUPAC-IUB Commission on Bioresin 22 was deprotected and neutralized in the usual way (Table I, chemical Nomenclature [J. Biol. Chem., 247, 977 (1972)] are used. Other abbreviations include DCC, dicylohexylcarbodiimide; DIEA, diisosteps 1-6) and then treated with 0.55 g (1.74 mmol) of 16 and 0.32 propylethylamine; TFA, trlfluoroacetic acid: DMB, 3,4-dimethyibenzyl. g (1.57 mmol) of DCC in CHzClz for 30 min. Following the reac(3) M. I. Blake, H. L. Crespie, V. Mohan, and J. J. Katz, J. Pharm. Sci., 50, tion, a ninhydrin test indicated that the reaction was ca. 94% com425 (1961). plete. A second coupling with 0.27 g (0.85 mmol) of 16 and 0.16 g (4) D. S. Berns, J. Am. Chem. SOC., 85, 1676 (1963); Biochemisfry,2, (0.80 mmol) of DCC gave >99.4% coupling as judged by the ninhy1377 (1963). (5) A. T. Blomquist, D. H. Rich, B. A. Carlson, G. A. Allen, V. J. Hruby, H. drin test. The Boc group was removed and the resin dried to give Takashima, L. L. Nangeroni, P. Glose, and V. du Vigneaud, Roc. Nafl. 4.87 g of the protected peptide-resin precursor of 4. After cleavage Acad. Sci,U.S.A., 64, 263 (1969). from the resin by ammonolysis, extraction of the peptide into (6) A. T. Blomquist, D. H. Rich, V. J. Hruby, L. L. Nangeroni, P. Glose, and DMF, and precipitation with water, 1.43 g of the protected nonaV. du Vigneaud, Proc. Natl. Acad. Sci. U.S.A., 61, 688 (1968). peptide amide was obtained, mp 220-225 "C. A 325-mg portion (7) V. du Vigneaud, J. D. Meador, M. F. Ferger, G. A. Allen, and A. T. Blomquist, Bioorg. Chem., 1, 123 (1971). was deprotected with sodium in liquid ammonia and oxidized in (8)V. J. Hruby in "Structure Activity Relationships of Protein and Peptide the manner described for compound 21. After purification by parHormones", M. Margoulies and F. C. Greenwood, Ed., Excerpta Medica, tition chromatography, the uncontaminated L diastereomer was Amsterdam, 1972, p 458. gel filtered on Sephadex G-25 using 0.2 N acetic acid as eluent, and (9) A. I. R. Brewster, J. A. Glasel, and V. J. Hruby, Proc. Nafl. Acad. Sci. (4a). lyophilized to give 40 mg of [l-hemi[~,/3-2H~]cystine]oxytocin U.S.A., 69, 1470 (1972). (IO) A. I. R. Brewster, V. J. Hruby, J. A. Glasel, and A. Tonelli, Biochemistry, It had [a1897 -22.2' ( c 0.5, 1 N HOAc). A sample of 4a was hydro12, 5294 (1973). lyzed in 6 N HC1 for 24 h at 110 OC. Amino acid analysis gave the (11) A. I. R. Brewster and V. J. Hruby, Proc. Natl. Acad. Sci. U.S.A., 70, following molar ratios: aspartic acid, 1.0; glutamic acid, 1.0; pro3806 (1973). line, 0.9; glycine, 1.O; half-cystine, 2.0; isoleucine, 1.0; leucine, 1.0; (12) F. A. Boveyin "Chemistry and Biology of Peptides", J. Meienhofer, Ed., tyrosine, 1.0. Development on TLC in systems A, B, and C gave Ann Arbor Science Publishers, Ann Arbor, Mich., 1972, p 3. [@OD

1358 J.Org. Chem., Vol. 41, No. 8, 1976

Kakinuma, Milavetz, and Rinehart

(13) A. I. R. Brewster. V. J. Hruby, A. F. Spatola, and F. A. Bovey, Biochemlstry, 12, 1643 (1973). (14) H. L. Crespie, U. Smith, L. Gajda, T. Tisue, and R. M. Ammeraai, Biochim. Biopbys. Acta, 256, 611 (1972), and references cited therein; H. L. Crespieand J. J. Katz, Pure Appl. Chem., 32,221 (1972). (15) 0. Jardetzky and N. G. WadeJardetzky, Annu. Rev. Biochem., 40, 605 (1971), and references cited therein. (16) 0. Jardetzky, Mol. Prop. Drug Recept., Ciba Found. Symp. 1970, 113 (1970). (17) J. S. Cohen, M. Feil, and 1. M. Chaiken, Biochim. Biophys. Acta, 236, 468 (1971). (18) 0. Jardetzky, H. Theilman, Y. Arata, J. L. Markley, and M. N. Williams, Cold Spring Harbor Symp. Quant. Biol., 36, 257 (1971), and references cited therein. (19) 0.Jardetzky, Proc. lnt. Conf. Stable /sot. Chem., Biol., Med., Ist, 1973, 99-102 (1973). (20) K. D. Kopple, A. Go, R. H. Logan, Jr., and S. Savrda, J. Am. Chem. SOC., 94, 973 (1972). (21) R. Schwyzer, C. Grathwohl, J.-P. Meraldi, A. Tun-Kyi, R. Vogel, and K. Wuthrich, Heiv. Chim. Acta, 55, 2545 (1972). (22)' A. F. Bradbury, A. S. V. Burgen, J. Feeney, G. C. K. Roberts, and D. G. Smyth, EBSLett., 42, 179 (1974). (23) C. Grathwohl, R. Schwyzer, A. Tun-Kyi, and K. Wuthrich, F€BS Lett., 29, 271 (1973). (24) J. A. Glasel, J. F. McKelvy, V. J. Hruby, and A. F. Spatola, Ann. N.Y. Acad. Sci., 222, 778 (1973). J. A. Glasel, V. J. Hruby, J. F. McKelvy, and A. F. Spatola, J. Mol. Biol., 79, 555 (1973). I. L. Schwartz in "The Chemistry of Polypeptides", P. G. Katsoyannis, Ed., Plenum Press, New York, N.Y., 1973, pp 297-313. R. Walter, I. L. Schwartz, J. H. Darneli, and D. W. Urry, Proc. Nati. Acad. Sci. U.S.A., 68,1355 (1971). R. Walter, Rec. Prog. Horm. Res., 28, 167(1972). V. J. Hruby in "Chemistry and Blochemistry of Amino Aclds, Peptides and Proteins", Vol. 3, B. Weinstein, Ed., Marcel Dekker, New York, N Y 1974. nn 1-188. (30) D. A.Corneiis, PhrD. Thesis, Cornell University, 1972. (31) 0. Yamashlro, Nature (London), 201, 76 (1964).

(32) D. Yamashiro, D. Gillessen, and V. du Vigneaud, J. Am. Chem. Soc., 88, 1310 (1966). (33) J. G. Atkinson, D. W. Cillis, and R. S. Stuart, Can. J. Chem., 47, 477 11969). (34) D. J. Pasto and G. R. Meyer, J. Org. Cbem., 33, 1257 (1968). (35) W. H. Grieve and K. F. Sporek, J. Chem. Educ., 43,381 (1966). (36) J. P. Greenstein and M. Winitz, "Chemistry of the Amino Acids", Vol. 3, Wiley, New York, N.Y., 1961, pp 1879-1928. (37) E. Schnabel. Justus Liebigs Ann. Chem., 702, 1881 (1967). (38) R. B. Merrifield, J. Am. Chem. SOC.,85,2149 (1963). (39) V. J. Hruby, L. E. Barstow, and T. Linhart, Anal. Chem., 44, 343 (1972). (40) E. Kaiser, R. L. Colescott, C. D. Bossinger. and P. I. Cook, Anal. Biochern., 34, 505 (1970). (41) W. Konig and R. Geiger, Chem. Ber., 103, 788 (1970). (42) R. H. Sifferd and V. du Vigneaud, J. Bioi. Chem., 108, 753 (1935). (43) M. Walti and D. B. Hope, Experientia, 29, 389 (1973). (44) D. B. Hope, V. Murti. and V. du Vigneaud, J. Biol. Chem., 237, 1563 (1962). (45) A. F. Spatola, D. A. Cornelius, V. J. Hruby, and A. T. Blomquist, J. Org, Chem., 39, 2207 (1974). (46) V. J. Hruby and M. E. Hadley in "Peptides: Chemistry, Structure and Biology", R. Walter and J. Meienhofer, Ed., Ann Arbor Science Publishers, Ann Arbor, Mlch., 1975. (47) D. A. Upson, V. J. Hruby, M. Blumenstein, and J.-P. Meraldl, unpublished results. (48) D. H. Spackman, W. H. Stein, and S. Moore, Anal. Chem., 30, 1190 (1958). (49) J. L. Wood and V. du Vigneaud, J. Bioi. Chem., 131, 267 (1939). (50) "Catalog of Infrared Spectrograms", Vol. 36, Sadtler Research Laboratories, Inc., Philadelphia, Pa., 1969, p 36546. (51) B. F. Gisen, Helv. Chim. Acta, 56, 1476 (1973). (52) K. W. Ehler, Ph.D. Thesis, University of Arizona, 1972. (53) K. Esko. S. Karlson, and J. Porath, Acta Chem. Scand., 22, 3342 (1968). (54) 0. H. Lowry, N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem., 193, 265 (1951). (55) M. Manning and V. du Vigneaud, J. Am. Chem. SOC.,87,3978 (1965). (56) L. C. Dorman, Biochem. Biopbys. Res. Commun., 80,318 (1974).

Carbon-13 Nuclear Magnetic Resonance Spectra of the Streptovaricins and Related Compounds192 Katsumi Kakinuma, Barry I. Milavetz, and Kenneth L. Rinehart, Jr.* School of Chemical Sciences, University of Illinois, Urbana, Illinois 61801 Received December 19,1975 Absorptions of the 40 carbon atoms of streptovaricin C have been assigned in the carbon magnetic resonance spectrum of that ansamycin antibiotic. Carbon absorptions of other streptovaricins,including streptovaricin D, whose biosynthesis has recently been studied, have also been assigned. Methods employed in assigning the individual carbons include off-resonance decoupling, specific proton decoupling, and comparison of the streptovaricins' spectra with spectra of one another, of compounds derived from the streptovaricins, and of model compounds. During the course of our extensive studies of the chemistry and biochemistry of the s t r e p t o ~ a r i c i n s ,members ~,~ of t h e ansamycin class of antibiotics: it has become necessary t o assign the chemical shifts of individual carbon atoms in their carbon magnetic resonance (13C NMR) spectra, both for studying the biosynthesis of streptovaricins2 and for the characterization of new, related compounds. In the present paper, we report these 13C NMR assignments.

Discussion Chemical shifts for the carbon atoms of streptovaricins A-E, G, and J (SVA-SVE, SVG,and SvJ, r e ~ p e c t i v e l y )and ~ for compounds derived from the streptovaricins (all of whose structures are shown in Figure 1) were determined on proton decoupled spectra and are summarized in Table I. Assignments were made by comparison of the spectra with proton off-resonance decoupled spectra, by single-frequency proton decoupling experiments, from standard chemical shift data, and by comparison with chemical shifts of model compounds, as discussed below. T h e most

abundant component of the complex, SvC, was employed as the reference compound for the assignments and the spectra of other compounds were compared with t h a t of SvC. This was especially valuable since SvC occupies a central position in the streptovaricin family (Figure 1). Thus, SVD is 14-deoxy-SvC, SVB is SvC 11-acetate, SvJ is SvC 7acetate, SVG is 6-hydroxy-SvC, and SVEis 7-oxo-7-deoxySvC. Streptovaricin A is SVG 11-acetate and SVFis O-demethyl-SvG-7-lactone. T h e spectrum of SvD is of special significance, since it is the component isolated in our biosynthetic studies employing 13C-labeled precursors.2 Carbon atom absorptions were divided initially into the groups shown in Table I according t o the number of attached hydrogen atoms (i.e., methyl, methylene, methine, quaternary carbons) by observations of the off-resonance decoupled spectra. Following this, some of the carbonsthe methoxy carbon, the quaternary aliphatic C-14, the methylenedioxy carbon, and the quinonoid carbonyl carbon (C-21)5-were assigned unambiguously from their offresonance multiplicities and characteristic chemical shifts6